I just about hate Tomb Raider right now.

Ergo, it’s a perfect time to announce my intentions to speedrun the game on PC.

Shortly, why I currently hate the game is I just came off of a 3 hour practice session where I couldn’t get the massive skywalking skip in Shantytown at all. This was after having gotten it twice in the span of less than an hour last night. To be clear though, it took me roughly 12.5 hours to get to the point of getting it the very first time, and then it was nearly 60 minutes until I was able to do it again.

That was last night. Tonight? Nothing.

But anyways.

Basically, my whole thing with Tomb Raider is simply the fact that I really like the game, and I’m just wanting to try this whole livestreaming thing out. That’s really where the whole speedrunning thing started. I just wanted to be able to start a stream and play the game from start to finish in one setting. I felt like that made it a more interesting viewing experience. It gave it a goal.

So I had decided that I would learn how to casually speedrun the game on PS4.

so why am I abandoning the PS4?

As odd as it is, this whole thing was kind of tipped just simply by the fact that I didn’t actually own Tomb Raider on the PS4. I had been blessed to get it for free from the library, and then further blessed to be able to renew it several times. But apparently, PS4s are getting much more popular in this area, so suddenly I’m greeted by a hold list that means that I probably won’t have the game back in my possession for another month or two. And that’s pretty much an eternity in speedrunning land. Especially considering I had just gotten started. Pretty much a momentum killer right there.

And in short, knowing that I was about to lose Tomb Raider started the ball rolling on getting me to consider getting a PC again. To cut to the chase, I got a PC.

So again, first major reason was that I don’t even own Tomb Raider on PS4.

Second major reason? The PS4 version occupies this limbo space where the framerate is too high to perform the tricks that work on PS3/xbox360/xboxone, but too low to perform the skywalking tricks on the PC.

I was already starting to learn this at the point that I was considering buying my own copy of Tomb Raider for PS4. So why spend massive amounts to buy it on PS4 when I might not even want to continue playing it on PS4.

So more specifically on the framerate, it apparently does variable framerates from 30-60 FPS. Skywalking tricks on PC appear to require about 150+ FPS. So no skywalking for you!

And then the 30-60 FPS is way too fast to pull off pause buffering. Or at least I base that on the fact that I personally have never successfully pause buffered on the PS4 version. But frankly, I may just not have the twitchiness required to pull off pause buffering at all.

why I’ll miss the PS4 version

I’d be lying if I didn’t mention that part of it is that, I think I actually have what it takes to be the world record holder on PS4. Why? I don’t think anyone else is running it on PS4. I win by default. Yay!

And on the PS4 version, I never had to worry about melting my CPU from disabling vsync (I’ll blog about how I fixed this soon; it’s basically underclocking; it’s also something I’d never have to know how to do on PS4).

Really though, I miss the controller. I never got RSI playing with the controller like I am now on mouse and keyboard.

I also was much better about standing up and moving around a bit any time there was something non crucial occurring (like a non-skippable cutscene), because hey, the controller can easily move with me.

but a keyboard and mouse is better for speedrunning

If I ever learn how to aim properly, well then there’s that. But really, the reason that I say a keyboard and mouse is better for speedrunning is repeatability.

I swear, if I had gone for the out of bounds (OOB) radio tower skip on the PC first, I wouldn’t have ever spent like 8 hours on it like I did on the PS4.

In brief, just after the part where Lara remarks about the snow and then some guy tries to throw you off a bridge, there’s a place where you can glitch into the side of a snowy embankment thingy and get out of bounds and avoid a whole bunch of fighting and just save a bunch of time in general.

So on the PS4, I was basically trying different strategies in terms of aim spots and controller movements. I’d find a specific spot that worked for me, get out of bounds, faff around a bit, then try to do it again using the same exact aim spot and movements.

And it wouldn’t flipping work.

It took me ages before I finally found a spot that worked for me most of the time.

I think there was just some bizarre combination of both controller knob positioning and timing that I was just not repeating each time. It was maddening. Even now, getting that skip feels more like intuition and less like a repeatable process.

Now compare that to the PC and keyboard/mouse (since you can play with controller on PC). Here, you have 4 possible direction keys. That’s it. No 360 degree controller knob that has god-knows how many possible combinations (e.g. there could be a big difference between pointing say 90 degrees to the right, and like 90.1 when it comes to certain tricks (I don’t know how precise you need to be though)).

But with keyboard and mouse, you can aim visually at something, and then you only have 4 possible direction keys to hit. And many of the tricks (so far anyhow) really only require simple sequences of a few keys.

Even better yet, you can have a keyboard overlay and know exactly what you pressed when a trick worked/didn’t work. It makes it hella useful for trying to repeat stuff. Or in my case, trying to diagnose failures.

Lara Skywalker

So what I’ve been literally working on for about the past 15 hours or so is the first massive skywalking glitch that occurs outside of Shantytown. I think it takes about 2 minutes or so if you do it right, and it saves about 23 minutes of gameplay based on everything that’s in brassmaster’s current world record console run.

And of course, as a total beginner, it also means saving yourself from learning a whole bunch of tricks that become immediately useless once you learn the skywalking trick in Shantytown.

For me, even though I’m zeroing in on this trick, I may well still go back and learn the old route.

For me, I think the biggest reason I’m learning this trick first, is simply that it’s the most compelling trick of the game for me. I’ve found that I just really enjoy faffing about out of bounds, and there’s a whole lot of out-of-bounds in this trick.

And it also feels special somehow. Like, if you get this trick, you’re immediately part of some elite group of glitch exploiters.

Finally I think it’s the fact that, if I can get this and the research base skywalking glitch, then I can basically learn nothing else and still get a time that allows me to easily sit down and do a run in one sitting.

Okay, scratch that, I keep forgetting about the fact that, if I learn the research base skywalking skip, then I also have to learn the cage puzzle skip because I won’t have the rope ascender.

So anyhow, I think all of that are reasons why I’m going for it right.

And hey, I’ve actually managed to get the first skywalking skip:

Twice even.

But the hella hella hella frustrating thing is that I spent 3 hours tonight trying and trying and trying and didn’t get it at all.

At this point, I can get out of bounds fairly quickly. I can get to the point where I start skywalking to the waterfall pretty quickly.

Then I’m failing to make it to the waterfall, I don’t know, maybe 50-75% of the time. My most common failing seems to be that I won’t get the necessary height to make it into the stream.

See, there’s this rubberbanding effect (which by the way seems to be the bit that’s framerate dependent in that, when I tested it at 60 FPS, I could skywalk, but I couldn’t rubberband upwards). But you need to get it a few times in order to successfully get high enough to make it into the stream. Fail to get high enough and you die.

So there’s my first huge sticking point so far.

Next is the final part of the skip (again with the skywalking). Half the time I fail by dying right off the bat (hitting some invisible spike in the ceiling??) and then half the time, I manage to get most of the way to the piece of land I need to get to, and then I end up literally about a meter or two from where I need to be. It’s a Wile E. Coyote type of moment, where it’s almost like Lara looks down and realizes there’s no floor beneath her and she plummets to her death.

Tonight, I was trying to experiment with looking at my footage from last night but it never really seemed to help. For one, I had only had enough time to drop markers in the footage, rather than what I think would be more helpful, which is to stich together all of the failures and all of the successes so I could view them back to back to back.

In fact, that’s still my intention. For instance, last night, when I finally bothered to look at my footage of how I successfully got through the waterfall, I then manage to get it successfully like 4 or 5 times in a row. And it taught me a method that I’m still using to this day (I use a different angle of entry to other people I’ve seen do the trick).

So I think it would be most helpful to go through all of my successes and all of my failures and look for common threads. Like, when I don’t get enough height, is it obvious what I did wrong? Is there a common thread? Similarly, is there any lessons I can learn from when I succeeded?

And frankly, I think it would also be good at this point to also track down as many samples from other people as I can find to look for common threads. But still, when it came to the waterfall ascension, it was my own pure dumb luck footage that taught me the most. Waiting for dumb luck is just torture though.

where I’m going from here

I am starting to reach a saturation point where I simply need to do anything else. I was excited going into tonight. I was supposed to spend tonight refining my methods and everything and getting to hopefully like a 50% success rate. Instead, I ended my 3 hour practice session with a 0% success rate on the entire skip.

On the plus side though, I did practice the initial fire arrows skip in Shantytown and I’m getting bizarrely more consistent with that (I think I’ve accidentally discovered the framerate sweet spot).

On the downside though, I’m discovering that door clips were far easier at 60 FPS. It’s even bad enough that some of them seem almost impossible at a higher rate (or at least, I have places where I’ve so far only gotten a clip with 60 FPS).

I’m wondering if I’m going to end up in some kind of hell where I have to deliberately turn vsync on and off because certain tricks are easier at the lower framerates. I mean, on the plus side, since I have zero intention of going for world record pace, it wouldn’t be the worst thing in the world to deck out to a menu to futz with vsync during an active run, but it just sucks. For one, I know me, and I know that I’ll attempt some trick several times before I realize that I’m at the wrong framerate to pull it off.

So yeah.

I was hoping to get the Shantyville skip consistently before I attempted an actual run. I feel like the worst thing would be to jump into a run and then simply be unable to move forward because I can’t get the trick.

I did realize that there’s a compromise. Given that I won’t be running for records, I could just cheat and, if I’m unable to get the trick after say 3 tries, I would just play a video of where I did get the trick (so people could know what was supposed to happen) and then load a checkpoint from when I did get the trick.

This would allow me to practice it in the context of an actual run, but give myself a fallback if I were simply unable to move past it.

Alternately, maybe I can find a trainer or something that will allow me to go out of bounds. Then I could enable it to get to where I’m supposed to get and then immediately disable it during a practice run.

Frankly, I need to get a trainer going anyway.

When you’re trying to figure out a specific trick, there’s nothing more frustrating than having to wait like 5 to 10 minutes before you even get another chance to try it. It also pretty much stops the hope for developing short-term muscle memory because there’s just way too much of a time gap between attempts.

But we’ll see how I’m feeling tomorrow. I may just take a day off. The only reason I haven’t so far (5 days in a row) is simply that the fact that I had been making forward progress. But I basically feel like I wasted all of tonight’s 3 hours. And since I was getting so frustrated, it’s not like I can even look at it as time-well-goofed-off.

I’m pretty sure I’m still in this for the long haul. And I think a sub 1:30 is in my future. Actually, I’m pretty curious to know what my total time will be with only the radio tower and the Shantyville skips.

But first, I have to get the stupid Shantyville skip consistently.

Argh.